EUVIMEDEUVIMED entdecken
Crossref

Scalable GPU Acceleration of Scalar Functions in Analytical Databases: Compilation, Benchmarking, and Optimization

Kaushik Rajan, Sampath Rajendra, Momin Al-Ghosien, Nicolas Bruno, Carlo Curino, Matteo Interlandi, Yinan Li, Lukas M. Maas, Craig Peeper, Surajit Chaudhuri, Johannes Gehrke

Proceedings of the VLDB Endowment · 01.03.2026 · 19 (7) · 1441-1454

DOI: 10.14778/3801059.3801061 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the VLDB Endowment, online erschienen am 14.07.2026. Verfasst von Kaushik Rajan und 10 weiteren Autor:innen. Die Arbeit nennt 62 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accelerating SQL query execution with GPUs is a central focus in database research. While prior systems have achieved notable speedups by offloading relational operators, the acceleration of the wide range of scalar functions that are supported by analytical engines remains unaddressed. Our analysis reveals that many scalar functions incur substantial computational overhead and often constitute the primary bottleneck in analytical queries on CPUs. This observation motivates a systematic exploration of the opportunities and challenges in accelerating scalar functions on GPUs. Unlike relational operators, which are few in number and standardized, production databases support hundreds of scalar functions. The absence of a standardized specification, combined with this diversity, renders manual GPU porting infeasible. To address this, we present an LLVM-MLIR-based compiler toolchain that automatically translates the CPU-based implementations of scalar functions from production databases into efficient GPU kernels, while preserving their original semantics. Our approach lifts scalar functions to a high-level intermediate representation, applies resource-optimizing transformations, and generates GPU assembly code, supporting all relevant data types, parameters, and database context variables. As existing benchmarks do not sufficiently stress test scalar functions in analytical queries, we introduce a variant of TPC-H that utilizes scalar functions while preserving the original query intent. Integrating our GPU kernels into a state-of-the-art GPU database system, we demonstrate substantial performance gains over a leading CPU database that uses slightly more expensive hardware: 7.6× on enhanced TPC-H and 6.4× on production queries, further widening the gap between GPU and CPU databases. The generated kernels deliver performance comparable to hand-optimized GPU implementations, establishing our approach as a scalable and practical solution for accelerating scalar functions on GPUs.

Zitieren

APA

Rajan, K., Rajendra, S., Al-Ghosien, M., Bruno, N., Curino, C., Interlandi, M., Li, Y., Maas, L. M., Peeper, C., Chaudhuri, S., & Gehrke, J. (2026). Scalable GPU Acceleration of Scalar Functions in Analytical Databases: Compilation, Benchmarking, and Optimization. Proceedings of the VLDB Endowment, 19(7), 1441-1454. https://doi.org/10.14778/3801059.3801061

Vancouver

Rajan K, Rajendra S, Al-Ghosien M, Bruno N, Curino C, Interlandi M, et al. Scalable GPU Acceleration of Scalar Functions in Analytical Databases: Compilation, Benchmarking, and Optimization. Proceedings of the VLDB Endowment. 2026;19(7):1441-1454. doi:10.14778/3801059.3801061

BibTeX
@article{Rajan2026,
  author = {Rajan, Kaushik and Rajendra, Sampath and Al-Ghosien, Momin and Bruno, Nicolas and Curino, Carlo and Interlandi, Matteo and Li, Yinan and Maas, Lukas M. and Peeper, Craig and Chaudhuri, Surajit and Gehrke, Johannes},
  title = {Scalable GPU Acceleration of Scalar Functions in Analytical Databases: Compilation, Benchmarking, and Optimization},
  journal = {Proceedings of the VLDB Endowment},
  year = {2026},
  volume = {19},
  number = {7},
  pages = {1441--1454},
  doi = {10.14778/3801059.3801061}
}
Kostenlos starten →